From: | Andrew Dunstan <andrew(at)dunslane(dot)net> |
---|---|
To: | Shachar Shemesh <shachar(at)shemesh(dot)biz> |
Cc: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>, oledb-devel(at)pgfoundry(dot)org, pgsql-hackers(at)postgresql(dot)org |
Subject: | Re: [Oledb-dev] Re: double precision error with pg linux server, but not with windows pg server |
Date: | 2007-05-20 04:52:12 |
Message-ID: | 464FD3FC.5010606@dunslane.net |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
Shachar Shemesh wrote:
>
> Even the original sentence does not describe the problem we're seeing
> here. It does not mention cross platform incompatibility.
>
That's why I suggested it should be improved.
>> The COPY docs are probably more correct: "The BINARY key word causes all
>> data to be stored/read as binary format rather than as text. It is
>> somewhat faster than the normal text mode, but a binary-format file is
>> less portable across machine architectures and PostgreSQL versions."
>>
>>
> Again, to me this sentence spells "you have a problem, we're not going
> to help you out, deal with it".
Eh? It spells out what we provide and what the limitations are. You
could put that spin on every documented limitation, if you were so inclined.
> This is especially true if what Tom said
> was true, that the text format does not maintain total precision. You
> are essentially telling the user "you cannot move your data reliably
> even between servers of the same version".
>
>
>
Since this is the exact mechanism used by pg_dump, we would surely have
been long since deafened by complaints if this were a problem of any
significance.
cheers
andrew
From | Date | Subject | |
---|---|---|---|
Next Message | Shachar Shemesh | 2007-05-20 05:43:08 | Re: Re: [Oledb-dev] double precision error with pg linux server, but not with windows pg server |
Previous Message | Tom Lane | 2007-05-20 04:33:35 | Re: Re: [Oledb-dev] double precision error with pg linux server, but not with windows pg server |